A system and method for machine-learning based atrial fibrillation detection are provided. A database is maintained that is operable to maintain a plurality of ecg features and annotated patterns of the features. At least one server is configured to: train a classifier based on the annotated patterns in the database; receive a representation of an ecg signal recorded by an ambulatory monitor recorder during a plurality of temporal windows; detect a plurality of the ecg features in at least some of the portions of the representation falling within each of the temporal windows; use the trained classifier to identify patterns of the ecg features within one or more of the portions of the ecg signal; for each of the portions, calculate a score indicative of whether the portion of the representation within that ecg signal is associated the patient experiencing atrial fibrillation; and take an action based on the score.
|
1. A system for machine-learning-based atrial fibrillation detection with the aid of a digital computer, comprising:
a database operable to maintain a plurality of electrocardiography (ecg) features and annotated patterns of the features, at least some of the patterns associated with atrial fibrillation;
at least one server interconnected to the database, the at least one server configured to:
train a classifier based on the annotated patterns in the database;
receive a representation of an ecg signal recorded by an ambulatory monitor recorder during a plurality of temporal windows;
detect a plurality of the ecg features in at least some of the portions of the representation falling within each of the temporal windows;
use the trained classifier to identify patterns of the ecg features within one or more of the portions of the ecg signal;
for each of the portions, calculate a value indicative of whether the portion of the representation within that ecg signal is associated the patient experiencing atrial fibrillation;
calculate a further value indicative of whether the portion of the representation within that ecg signal is associated with the patient not experiencing atrial fibrillation;
compare the further value to the value;
determine that the portion of the ecg signal is associated with the patient experiencing atrial fibrillation based on the comparison; and
take an action based on the determination that the portion of the ecg signal is associated with the patient experiencing atrial fibrillation.
10. A method for machine-learning-based atrial fibrillation detection with the aid of a digital computer, comprising:
maintaining in a database a plurality of electrocardiography (ecg) features and annotated patterns of the features, at least some of the patterns associated with atrial fibrillation;
training by an at least one server connected to the database a classifier based on the annotated patterns in the database;
receiving by the at least one server a representation of an ecg signal recorded by an ambulatory monitor recorder during a plurality of temporal windows;
detecting by the at least one server a plurality of the ecg features in at least some of the portions of the representation falling within each of the temporal windows;
using by the at least one server the trained classifier to identify patterns of the ecg features within one or more of the portions of the ecg signal;
for each of the portions, calculating by the at least one server a value indicative of whether the portion of the representation within that ecg signal is associated the patient experiencing atrial fibrillation;
calculating by the at least one server a further value indicative of whether the portion of the representation within that ecg signal is associated with the patient not experiencing atrial fibrillation;
comparing the further value to the score;
determining that the portion of the ecg signal is associated with the patient experiencing atrial fibrillation based on the comparison;
taking by the at least one server an action based on the determination that the portion of the ecg signal is associated with the patient experiencing atrial fibrillation.
2. A system for machine-learning-based atrial fibrillation detection according to
obtain training data comprising a plurality of the ecg features and a plurality of patterns of the ecg features; and
obtain annotations of patterns of the ecg features in the training data, wherein the training of the classifier is based on the annotations.
3. A system for machine-learning-based atrial fibrillation detection according to
test an accuracy of the trained classifier and perform further training based on a result of the test.
4. A system for machine-learning-based atrial fibrillation detection according to
5. A system for machine-learning-based atrial fibrillation detection according to
6. A system for machine-learning-based atrial fibrillation detection according to
generate a matrix with the identified features and the patterns; and
generate at least one matrix with weights for the identified features and patterns, wherein the value and the further value are calculated using the weight matrix.
7. A system for machine-learning-based atrial fibrillation detection according to
8. A system for machine-learning-based atrial fibrillation detection according to
9. A system for machine-learning-based atrial fibrillation detection according to
perform a noise filtering of at least some of the portions of the ecg signal prior to identification of the ecg features.
11. A method for machine-learning-based atrial fibrillation detection according to
obtaining training data comprising a plurality of the ecg features and a plurality of patterns of the ecg features; and
obtaining annotations of patterns of the ecg features in the training data, wherein the training of the classifier is based on the annotations.
12. A method for machine-learning-based atrial fibrillation detection according to
test an accuracy of the trained classifier and performing further training based on a result of the test.
13. A method for machine-learning-based atrial fibrillation detection according to
14. A method for machine-learning-based atrial fibrillation detection according to
15. A method for machine-learning-based atrial fibrillation detection according to
generating a matrix with the identified features and the patterns; and
generating at least one matrix with weights for the identified features and patterns, wherein the value and the further value are calculated using the weight matrix.
16. A method for machine-learning-based atrial fibrillation detection according to
17. A method for machine-learning-based atrial fibrillation detection according to
18. A method for machine-learning-based atrial fibrillation detection according to
performing a noise filtering of at least some of the portions of the ecg signal prior to identification of the ecg features.
|
|||||||||||||||||||||||||||||
This present non-provisional patent application is a continuation-in-part of U.S. Pat. No. 10,165,946, issued Jan. 1, 2019, which is a continuation-in-part of U.S. Pat. No. 9,433,367, issued Sep. 6, 2016; which is a continuation-in-part of U.S. Pat. No. 9,545,204, issued Jan. 17, 2017, and a continuation-in-part of U.S. Pat. No. 9,730,593, issued Aug. 15, 2017; and which further claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent application, Ser. No. 61/882,403, filed Sep. 25, 2013, the disclosures of which are incorporated by reference; this non-provisional application also claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent application, Ser. No. 62/591,715, filed Nov. 28, 2017, the disclosures of which is incorporated by reference.
This application relates in general to electrocardiographic monitoring and, in particular, to a system and method for machine-learning-based atrial fibrillation detection.
An electrocardiogram (ECG) records electrical potentials in the heart using a standardized set format 12-lead configuration to record cardiac electrical signals from well-established chest locations. Electrodes are placed on the skin over the anterior thoracic region of the body to the lower right and to the lower left of the sternum, on the left anterior chest, and on the limbs. The recorded cardiac electrical activity, represented by PQRSTU waveforms, can be interpreted to derive heart rate and physiology, where each P-wave represents atrial electrical activity and the QRSTU components represent ventricular electrical activity.
An ECG is a snapshot of heart function, typically recorded over 12 seconds, that can help diagnose rate and regularity of heartbeats, effect of drugs or implantable cardiac devices, and whether a patient has heart disease. ECGs are limited to recording those heart-related aspects present at the time of recording. Thus, an ECG only provides a partial picture and can be insufficient for complete patient diagnosis of many cardiac disorders. Sporadic conditions that may not show up during a spot ECG recording, including fainting or syncope; rhythm disorders, such as tachyarrhythmias and bradyarrhythmias; asystolic episodes; and other cardiac and related disorders, require other means of diagnosis.
Diagnostic efficacy of cardiac rhythm disorders in particular can be improved, when appropriate, through long-term extended ECG monitoring. Recording cardiac physiology over an extended period can be challenging, yet is often essential to enabling a physician to identify events of potential concern. Although a 30-day observation period is considered the “gold standard” of ECG monitoring, achieving 30-day coverage has proven unworkable because conventional ambulatory ECG monitoring systems are arduous to employ, cumbersome to wear, and excessively costly. Nevertheless, if a patient's ECG could be recorded in an ambulatory setting over long periods of time, thereby allowing the patient to engage in activities of daily living, the chances of acquiring meaningful information and capturing an abnormal cardiac rhythm event becomes more likely to be achieved.
For instance, the long-term wear of ECG electrodes is complicated by skin irritation and the inability of conventional ECG electrodes to maintain continual skin contact after a day or two. Moreover, time, dirt, moisture, and other environmental contaminants, as well as perspiration, skin oil, and dead skin cells from the patient's body, can get between an ECG electrode, the non-conductive adhesive used to adhere the ECG electrode, and the skin's surface. These factors adversely affect electrode adhesion and the quality of cardiac signal recordings. Moreover, physical movement and clothing impart compressional, tensile, and torsional forces on electrode contact points decreasing signal quality, especially over long recording times. In addition, an inflexibly fastened ECG electrode is prone to dislodgement that often occurs unbeknownst to the patient, making the ECG recordings worthless. Further, some patients may have skin conditions, such as itching and irritation, aggravated by the wearing of most ECG electrodes. A patient may have to periodically remove or replace electrodes during a long-term monitoring period, whether to replace a dislodged electrode, reestablish better adhesion, alleviate itching or irritation, allow for cleansing of the skin, allow for showering and exercise, or for other purpose. Such replacement or slight alteration in electrode location can interfere with the goal of recording the ECG signal for long periods of time. Finally, such recording devices are often ineffective at recording atrial electrical activity, which is critical in the accurate diagnosis of heart rhythm disorders, because of the use of traditional ECG recording electronics or due to the location of the monitoring electrodes far from the origin of the atrial signal, for instance, the P-wave.
Conventionally, Holter monitors are widely used for long-term extended ECG monitoring, typically, for only 24-48 hours. A typical Holter monitor is a wearable and portable version of an ECG that includes cables for each electrode placed on the skin and a separate battery-powered ECG recorder. Similar to standard in-clinic ECG practice, the cable and electrode combination (or leads) are placed in the anterior thoracic region. The duration of a Holter monitoring recording depends on the sensing and storage capabilities of the monitor, as well as battery life. A “looping” Holter (or event) monitor can operate for a longer period of time by overwriting older ECG tracings, thence “recycling” storage in favor of extended operation, yet at the risk of losing crucial event data. Holter monitors remain cumbersome, expensive and typically for prescriptive use only, which limits their usability. Further, the skill required to properly place the ECG leads on the patient's chest hinders or precludes a patient from replacing or removing the electrodes.
The ZIO XT Patch and ZIO Event Card devices, manufactured by iRhythm Tech., Inc., San Francisco, Calif., are wearable stick-on monitoring devices that are typically worn on the upper left pectoral region to respectively provide continuous and looping ECG recordings. The location is used to simulate surgically implanted monitors. These devices are prescription-only for single patient use. The ZIO XT Patch device is limited to 14-day monitoring, while the ZIO Event Card device's electrodes can be worn for up to 30 days. The ZIO XT Patch device combines electronic recordation components, including battery, and physical electrodes into a unitary assembly that adheres to the skin. The ZIO XT Patch device uses adhesive strong enough to support the weight of both the monitor and the electrodes over an extended period of time. During monitoring, the battery is continually depleted and can potentially limit overall monitoring duration. The ZIO Event Card device is a form of downsized Holter monitor with a recorder component that must be removed during activities that could damage the non-waterproof electronics. These patches have a further limitation because of a small inter-electrode distance coupled to its designed location of application, high on the left chest. The electrical design of the ZIO patch and its location make recording high quality atrial signals (P-waves) difficult, as the location is relatively far from the origin of these low amplitude signals. As well, the location is suboptimal for identification of these signals. Furthermore, this patch is problematical for woman by being placed in a location that may limit signal quality, especially in woman with large breasts or bosoms. Both ZIO devices represent compromises between length of wear and quality of ECG monitoring, especially with respect to ease of long term use, female-friendly fit, and quality of atrial (P-wave) signals.
Personal ambulatory monitoring, both with smartphones or via adjuncts to smartphones, such as with a wirelessly-connected monitor or activity tracker, of varying degrees of sophistication and interoperability, have become increasingly available. For instance, McManus et al., “A Novel Application for the Detection of an Irregular Pulse using an iPhone 4S in Patients with Atrial Fibrillation,” Vol. 10(3), pp. 315-319 (March 2013), the disclosure of which is incorporated by reference, discloses obtaining pulsatile time series recordings before and after cardioversion using the digital camera built into a smartphone. An algorithm implemented as an app executed by the smartphone analyzed recorded signals to accurately distinguish pulse recordings during atrial fibrillation from sinus rhythm, although such a smartphone-based approach provides non-continuous observation and would be impracticable for long term physiological monitoring. Further, the smartphone-implemented app does not provide continuous recordings, including the provision of pre-event and post-event context, critical for an accurate medical diagnosis that might trigger a meaningful and serious medical intervention. In addition, a physician would be loath to undertake a surgical or serious drug intervention without confirmatory evidence that the wearer in question was indeed the subject of the presumed rhythm abnormality. Validation of authenticity of the rhythm disorder for a specified patient takes on critical legal and medical importance.
The AliveCor heart monitor, manufactured by AliveCor, Inc., San Francisco, Calif., provides a non-continuous, patient-triggered event monitor, which is worn on the fingertip. Heart rate is sensed over a single lead (comparable to Lead I on a conventional ECG) and recorded by an app running on a smartphone, such as an iOS operating system-based smartphone, such as the iPhone, manufactured by Apple Inc., Cupertino, Calif., or an Android operating system-based smartphone, manufactured and offered by various companies, including Google Inc., Mountain View, Calif.; Samsung Electronics Co., Ltd., Suwon, S. Korea; Motorola Mobility LLC, a subsidiary of Google Inc., Libertyville, Ill.; and LG Electronics Inc., Seoul, S. Korea. The Android operating system is also licensed by Google Inc. The app can send the data recorded by an AliveCor heart monitor from the smartphone to healthcare providers, who ultimately decide whether to use the data for screening or diagnostic purposes. Furthermore, as explained supra with respect to the McManus reference, none of these devices provides the context of the arrhythmia, as well as the medico-legal confirmation that would otherwise allow for a genuine medical intervention.
Similarly, adherents to the so-called “Quantified Self” movement combine wearable sensors and wearable computing to self-track activities of their daily lives. The Fitbit Tracker, manufactured by Fitbit Inc., San Francisco, Calif.; the Jawbone UP, manufactured by Jawbone, San Francisco, Calif.; the Polar Loop, manufactured by Polar Electro, Kempele, Finland; and the Nike+ FuelBand, manufactured by Nike Inc., Beaverton, Oreg., for instance, provide activity trackers worn on the wrist or body with integrated fitness tracking features, such as a heart rate monitor and pedometer to temporally track the number of steps taken each day with an estimation calories burned. The activity tracker can interface with a smartphone or computer to allow a wearer to monitor their progress towards a fitness goal. These activity trackers are accessories to smartphones, including iOS operating system-based smartphones, Android operating system-based smartphones, and Windows Phone operating-system based smartphones, such as manufactured by Microsoft Corporation, Redmond, Wash., to which recorded data must be offloaded for storage and viewing.
The features of activity trackers can also be increasingly found in so-called “smart” watches that combine many of the features of activity trackers with smartphones. Entire product lines are beginning to be offered to provide a range of fitness- and health-tracking solutions. As one example, Samsung Electronics Co., Ltd., offers a line of mobile products with fitness-oriented features, including the Galaxy S5 smartphone, which incorporates a biometric fingerprint reader and heart rate monitor; the Gear 2 smart watch, which also incorporates a heart rate monitor; and the Gear Fit wearable device, which incorporates a heart rate monitor, real time fitness coaching, and activity tracker. The Galaxy S5 smartphone's heart rate monitor is not meant for continuous tracking, while the both the Gear 2 smart watch and Gear Fit wearable device must be paired with a smartphone or computer to offload and view the recorded data.
With all manner of conventional “fitness-oriented” device, whether smartphone, smart watch, or activity tracker, quantified physiology is typically tracked for only the personal use of the wearer. Monitoring can be either continuous or non-continuous. The wearer must take extra steps to route recorded data to a health care provider; thus, with rare exception, the data is not time-correlated to physician-supervised monitoring nor validated. Furthermore, the monitoring is strictly informational and medically-significant events, such as serious cardiac rhythm disorders, including tachyarrhythmias and bradyarrhythmias, while potentially detectable, are neither identified nor acted upon.
In today's medical and legal environment, a mobile device, such as a smartphone, provides information that cannot be translated into data that triggers surgery or drug therapy by a physician. In the case of a smartphone detecting a fast heartbeat, for example, such a detection and the information on the smartphone would neither be identified as truly related to the patient in question or would be deemed sufficient for subjecting a patient to surgery or potentially toxic drug therapy. Thus, such data that is available today is not actionable in a medically therapeutic relevant way. To make such data actionable, one must have recorded data that allows a specific rhythm diagnosis, and a vague recording hinting that something may be wrong with the heart will not suffice. Further, the recorded data must not only identify the heart-related event of concern, but the signals before and after the event, which provides critical medical information for a physician to diagnose the disorder specifically. Finally, the recorded data must be made certifiable, so that the relationship of the recorded data to the patient that the physician is seeing is clear and appropriately identifiable as an event originating in the patient being examined. Establishing this relationship of data-to-patient has become a legally mandatory step in providing medical care, which heretofore has been impracticable insofar as one cannot merely rely upon a smartphone to provide legally sufficient identification of an abnormality with actionable data such that a patient undergoes a serious medical intervention.
Even once ECG monitoring data of sufficient length and quality to serve as a basis for a diagnosis is obtained, further challenges exist in providing an efficient and actionable interpretation of the data. While a physician may personally perform an over-read of the data over the entire length of the monitoring manually, such an over-read consumes a significant amount of time that will slow down delivering any necessary treatment to the patient. While automated interpretation techniques exist, such techniques still have significant limitations in recognizing certain types of cardiac conditions. For example, such techniques may have trouble recognizing atrial fibrillation (“AF”), a condition characterized by a rapid, irregular, beating of the atrium. AF is associated with an increased risk of stroke and heart failure, and is thus an important condition to timely diagnose. Most automated algorithms trying to diagnose AF focus on individual features of the ECG waves, such as particularly focusing on distances between R-waves and comparing them to predefined thresholds. Such an approach ignores other manifestations of AF and may lead to misdiagnosis AF is initially diagnosed by an absence of organized P-waves 11 and confirmed by erratic ventricular rates that manifest in an ECG R-R interval plot as a cloud-like pattern of irregular R-R intervals due to an abnormal conduction of impulses to the ventricles.
Therefore, a need remains for an extended wear continuously recording ECG monitor practicably capable of being worn for a long period of time in both men and women and capable of recording high quality atrial and ventricular signals reliably.
A further need remains for facilities to integrate wider-ranging physiological and “life tracking”-type data into long-term ECG and physiological data monitoring coupled with an onboard ability to cascade into the medical records and to the medical authorities appropriate medical intervention upon detection of a condition of potential medical concern.
A still further need remains for a flexible way to detect atrial fibrillation based on results of an ECG monitoring results that allows to take into account diverse manifestations of atrial fibrillations on an ECG trace.
Physiological monitoring can be provided through a wearable monitor that includes two components, a flexible extended wear electrode patch and a removable reusable monitor recorder. The wearable monitor sits centrally (in the midline) on the patient's chest along the sternum oriented top-to-bottom. The placement of the wearable monitor in a location at the sternal midline (or immediately to either side of the sternum), with its unique narrow “hourglass”-like shape, benefits long-term extended wear by removing the requirement that ECG electrodes be continually placed in the same spots on the skin throughout the monitoring period. Instead, the patient is free to place an electrode patch anywhere within the general region of the sternum, the area most likely to record high quality atrial signals or P-waves. Moreover, the wearable monitor is worn in such a location that is comfortable to woman and allows wear during activity.
In a further embodiment, the wearable monitor can interoperate wirelessly with other wearable physiology monitors and activity sensors and with mobile devices, including so-called “smartphones,” as well as with personal computers and tablet or handheld computers, to download monitoring data either in real-time or in batches. Where a wearable physiology monitor, activity sensor, or mobile device worn or held by the patient includes the capability to sense cardiac activity, particularly heart rate, or other physiology, an application executed by the monitor, sensor, or device can trigger the dispatch of a medically-actionable wearable monitor to the patient upon detecting potentially medically-significant events, such as cardiac rhythm disorders, including tachyarrhythmias and bradyarrhythmias. Upon receipt of the wearable monitor, the patient can use the sensor or device, if appropriately equipped with photographic, fingerprint or thumbprint, or other recording features, to physically record the placement and use of the wearable monitor, thereby facilitating the authentication of the data recorded by the wearable monitor. Finally, the monitor recorder can also be equipped with a wireless transceiver to either provide data or other information to, or receive data or other information from, an interfacing wearable physiology monitor, activity sensor, or mobile device for relay to an external system or further device, such as a server, analysis, or for further legal validation of the relationship of the monitor to the patient, or for other purpose.
The monitoring patch is especially suited to the female anatomy. The narrow longitudinal midsection can fit nicely within the intermammary cleft of the breasts without inducing discomfort, whereas conventional patch electrodes are wide and, if adhesed between the breasts, particularly hypertrophic or pendulous breasts, would cause chafing, irritation, frustration, and annoyance, leading to low patient compliance and possibly premature removal of the monitoring patch. Furthermore, such conventional patches do not have the electrical design and signal processing that would allow recording of the P-wave, given the close spacing of the bipolar electrodes.
The foregoing aspects enhance ECG monitoring performance and quality, facilitating long-term ECG recording, critical to accurate arrhythmia diagnosis.
In addition, the foregoing aspects enhance comfort in women (and certain men), but not irritation of the breasts, by placing the monitoring patch in the best location possible for optimizing the recording of cardiac signals from the atrium, another feature critical to proper arrhythmia diagnosis.
Finally, the foregoing aspects as relevant to monitoring are equally applicable to recording other physiological data, such as heart rate, temperature, blood pressure, respiratory rate, blood pressure, blood sugar (with appropriate subcutaneous probe), oxygen saturation, minute ventilation, as well as other measures of body chemistry and physiology.
Further, instances of atrial fibrillation experienced during a cardiac monitoring can be flexibly detected using machine learning. ECG training data is obtained and ECG trace features in the data are identified. Patterns of ECG features in the training data are annotated as either being indicative of a patient having experienced atrial fibrillation or not indicative of a patient having experienced atrial fibrillation. An adaptive atrial fibrillation detector is trained on the annotated patterns and the trained classifier is used to recognize patterns of ECG features indicative of atrial fibrillation in newly-acquired ECG monitoring data. Due to being trained on patterns of ECG features, with more training being possible as more patterns are revealed, the adaptive atrial fibrillation detector can detect more manifestations of atrial fibrillation recorded during an ECG monitoring that would be available through simple use of predefined parameters, such as distances between successive R-waves. Once the atrial fibrillation detection is confirmed by an appropriate medical professional, the detection can serve as a basis for initiating an appropriate treatment of the patient or another action.
In one embodiment, a system and method for machine-learning based atrial fibrillation detection with the aid of a digital computer are provided. A database is maintained that is operable to maintain a plurality of electrocardiography (ECG) features and annotated patterns of the features, at least some of the patterns associated with atrial fibrillation. At least one server is interconnected to the database, the at least one server configured to: train a classifier based on the annotated patterns in the database; receive a representation of an ECG signal recorded by an ambulatory monitor recorder during a plurality of temporal windows; detect a plurality of the ECG features in at least some of the portions of the representation falling within each of the temporal windows; use the trained classifier to identify patterns of the ECG features within one or more of the portions of the ECG signal; for each of the portions, calculate a score indicative of whether the portion of the representation within that ECG signal is associated the patient experiencing atrial fibrillation; and take an action based on the score.
Still other embodiments will become readily apparent to those skilled in the art from the following detailed description, wherein are described embodiments by way of illustrating the best mode contemplated. As will be realized, other and different embodiments are possible and the embodiments' several details are capable of modifications in various obvious respects, all without departing from their spirit and the scope. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.
Physiological monitoring can be provided through a wearable monitor that includes two components, a flexible extended wear electrode patch and a removable reusable monitor recorder.
The placement of the wearable monitor 12 in a location at the sternal midline 16 (or immediately to either side of the sternum 13) significantly improves the ability of the wearable monitor 12 to cutaneously sense cardiac electric signals, particularly the P-wave (or atrial activity) and, to a lesser extent, the QRS interval signals in the ECG waveforms that indicate ventricular activity, while simultaneously facilitating comfortable long-term wear for many weeks. The sternum 13 overlies the right atrium of the heart and the placement of the wearable monitor 12 in the region of the sternal midline 13 puts the ECG electrodes of the electrode patch 15 in a location better adapted to sensing and recording P-wave signals than other placement locations, say, the upper left pectoral region or lateral thoracic region or the limb leads. In addition, placing the lower or inferior pole (ECG electrode) of the electrode patch 15 over (or near) the Xiphoid process facilitates sensing of ventricular activity and provides superior recordation of the QRS interval.
When operated standalone, the monitor recorder 14 of the extended wear electrocardiography and physiological sensor monitor 12 senses and records the patient's ECG data into an onboard memory. In addition, the wearable monitor 12 can interoperate with other devices.
Upon retrieving stored ECG monitoring data from a monitor recorder 14, middleware first operates on the retrieved data to adjust the ECG capture quality, as necessary, and to convert the retrieved data into a format suitable for use by third party post-monitoring analysis software, as further described infra with reference to
A client-server model could be used to employ a server 122 to remotely interface with the download station 125 over the network 121 and retrieve the formatted data or other information. The server 122 executes a patient management program 123 (“Mgt”) or similar application that stores the retrieved formatted data and other information in a secure database 124 cataloged in that patient's EMRs 134. In addition, the patient management program 123 could manage a subscription service that authorizes a monitor recorder 14 to operate for a set period of time or under pre-defined operational parameters.
The patient management program 123, or other trusted application, also maintains and safeguards the secure database 124 to limit access to patient EMRs 134 to only authorized parties for appropriate medical or other uses, such as mandated by state or federal law, such as under the Health Insurance Portability and Accountability Act (HIPAA) or per the European Union's Data Protection Directive. For example, a physician may seek to review and evaluate his patient's ECG monitoring data, as securely stored in the secure database 124. The physician would execute an application program 130 (“Pgm”), such as a post-monitoring ECG analysis program, on a personal computer 129 or other connectable computing device, and, through the application 130, coordinate access to his patient's EMRs 134 with the patient management program 123. Other schemes and safeguards to protect and maintain the integrity of patient EMRs 134 are possible. The patient management program 123 can also manage access to the EMRs by a variety of parties that are authorized to access to modify the EMRs, as further described in “ . . . ,” which is hereby incorporated by reference.
Prior to Storing the Results of the Monitoring into the EMRs, the EMRs
The wearable monitor 12 can interoperate wirelessly with other wearable physiology monitors and activity sensors 131, such as activity trackers worn on the wrist or body, and with mobile devices 133, including smart watches and smartphones. Wearable physiology monitors and activity sensors 131 encompass a wide range of wirelessly interconnectable devices that measure or monitor a patient's physiological data, such as heart rate, temperature, blood pressure, respiratory rate, blood pressure, blood sugar (with appropriate subcutaneous probe), oxygen saturation, minute ventilation, and so on; physical states, such as movement, sleep, footsteps, and the like; and performance, including calories burned or estimated blood glucose level. The physiology sensors in non-wearable mobile devices, particularly smartphones, are generally not meant for continuous tracking and do not provide medically precise and actionable data sufficient for a physician to prescribe a surgical or serious drug intervention; such data can be considered screening information that something may be wrong, but not data that provides the highly precise information that may allow for a surgery, such as implantation of a pacemaker for heart block or a defibrillator for ventricular tachycardia, or the application of serious medications, like blood thinners for atrial fibrillation or a cardiac ablation procedure. Such devices, like smartphones, are better suited to pre- and post-exercise monitoring or as devices that can provide a signal that something is wrong, but not in the sufficient detail and validation to allow for medical action. Conversely, medically actionable wearable sensors and devices sometimes provide continuous recording for relatively short time periods, but must be paired with a smartphone or computer to offload and evaluate the recorded data, especially if the data is of urgent concern.
Wearable physiology monitors and activity sensors 131, also known as “activity monitors,” and to a lesser extent, “fitness” sensor-equipped mobile devices 133, can trace their life-tracking origins to ambulatory devices used within the medical community to sense and record traditional medical physiology that could be useful to a physician in arriving at a patient diagnosis or clinical trajectory, as well as from outside the medical community, from, for instance, sports or lifestyle product companies who seek to educate and assist individuals with self-quantifying interests. Data is typically tracked by the wearable physiology monitors or activity sensors 131 and mobile device 133 for only the personal use of the wearer. The physiological monitoring is strictly informational, even where a device originated within the medical community, and the data is generally not time-correlated to physician-supervised monitoring. Importantly, medically-significant events, such as cardiac rhythm disorders, including tachyarrhythmias, like ventricular tachycardia or atrial fibrillation, and bradyarrhythmias, like heart block, while potentially detectable with the appropriate diagnostic heuristics, are neither identified nor acted upon by the wearable physiology monitors and activity sensors 131 and the mobile device 133.
Frequently, wearable physiology monitors and activity sensors 131 are capable of wirelessly interfacing with mobile devices 133, particularly smart mobile devices, including so-called “smartphones” and “smart watches,” as well as with personal computers and tablet or handheld computers, to download monitoring data either in real-time or in batches. The wireless interfacing of such activity monitors is generally achieved using transceivers that provide low-power, short-range wireless communications, such as Bluetooth, although some wearable physiology monitors and activity sensors 131, like their mobile device cohorts, have transceivers that provide true wireless communications services, including 4G or better mobile telecommunications, over a telecommunications network. Other types of wireless and wired interfacing are possible.
Where the wearable physiology monitors and activity sensors 131 are paired with a mobile device 133, the mobile device 133 executes an application (“App”) that can retrieve the data collected by the wearable physiology monitor and activity sensor 131 and evaluate the data to generate information of interest to the wearer, such as an estimation of the effectiveness of the wearer's exercise efforts. Where the wearable physiology monitors and activity sensors 131 has sufficient onboard computational resources, the activity monitor itself executes an app without the need to relay data to a mobile device 133. Generally, such more computationally-capable wearable physiology monitors and activity sensors are also equipped with wireless communications services transceivers, such as found in some smart watches that combine the features of activity monitors with mobile devices. Still other activity monitor and mobile device functions on the collected data are possible.
In a further embodiment, a wearable physiology monitor, activity sensor 131, or mobile device 133 worn or held by the patient 10, or otherwise be used proximal to the patient's body, can be used to first obtain and then work collaboratively with a more definitive monitor recorder 14 to enable the collection of physiology by the monitor recorder 14 before, during, and after potentially medically-significant events. The wearable physiology monitor, activity sensor 131, or mobile device 133 must be capable of sensing cardiac activity, particularly heart rate or rhythm, or other types of physiology or measures, either directly or upon review of relayed data. Where the wearable physiology monitor or activity sensor 131 is paired with a mobile device 133, the mobile device 133 serves as a relay device and executes an application that will trigger the dispatch of a monitor recorder 14 to the patient 10 upon detecting potentially medically-significant events in the data provided by the paired activity monitor, such as cardiac rhythm disorders, including tachyarrhythmias and bradyarrhythmias. If the mobile device 133 is itself performing the monitoring of the patient's physiology, the mobile device 133 executes an application that will trigger the dispatch of a monitor recorder 14 to the patient 10 in near-real time upon detecting potentially medically-significant events, thereby avoiding the delay incurred by data relay from an activity monitor. Finally, if the wearable physiology monitor or activity sensor 131 has sufficient onboard computational resources and also is equipped with a wireless communications services transceiver, the wearable physiology monitor or activity sensor 131 effectively becomes the mobile device 133 and executes an application that will trigger the dispatch of a monitor recorder 14 to the patient 10 in near-real time upon detecting potentially medically-significant events without the need to first interface with a mobile device 133. Still other configurations of the detection app are possible.
The act of triggering the dispatch of a monitor recorder 14 represents the first step in a cascade of possible medical interventions of potentially increasing seriousness and urgency. Sensors 131 and devices 133 are generally not capable of detecting and recording medically precise and actionable data, whereas, as a device designed for extended wear, the monitor recorder 14 continually monitors the patient's physiology over a long time period and will capture any medically-actionable data leading up to, throughout the occurrence of, and following an event of potential medical concern.
The monitoring data recorded by the monitor recorder 14 can be uploaded directly into the patient's EMRs 134, either by using a mobile device 133 as a conduit for communications with a server 122 coupled to a secure database 124 within which the patient's EMRs 134 are stored, or directly to the server 122, if the monitor recorder 14 is appropriately equipped with a wireless transceiver or similar external data communications interface, as further described infra. Thus, the data recorded by the monitor recorder 14 would directly feed into the patient's EMRs 134, thereby allowing the data to be made certifiable for immediate use by a physician or healthcare provider. No intermediate steps would be necessary when going from cutaneously sensing cardiac electric signals and collecting the patient's physiology using a monitor recorder 14 to presenting that recorded data to a physician or healthcare provider for medical diagnosis and care. The direct feeding of data from the monitor recorder 14 to the EMRs 134 clearly establishes the relationship of the data, as recorded by the monitor recorder 14, to the patient 10 that the physician is seeing and appropriately identifies any potentially medically-significant event recorded in the data as originating in the patient 10 and nobody else. Based on the monitoring data, physicians and healthcare providers can rely on the data as certifiable and can directly proceed with determining the appropriate course of treatment for the patient 10, including undertaking further medical interventions as appropriate. In a further embodiment, the server 122 can evaluate the recorded data, as fed into the patient's EMRs 134, to refer the patient 10 for medical care to a general practice physician or medical specialist, for instance, a cardiac electrophysiologist referral from a cardiologist when the recorded data indicates an event of sufficient potential severity to warrant the possible implantation of a pacemaker for heart block or a defibrillator for ventricular tachycardia. Other uses of the data recorded by the monitor recorder 14 are possible.
For instance, a patient 10 who has previously suffered heart failure is particularly susceptible to ventricular tachycardia following a period of exercise or strenuous physical activity. A wearable sensor 131 or device 133 that includes a heart rate monitor would be able to timely detect an irregularity in heart rhythm. The application executed by the sensor 131 or device 133 allows those devices to take action by triggering the dispatch of a monitor recorder 14 to the patient 10, even though the data recorded by the sensor 131 or device 133 is itself generally medically-insufficient for purposes of diagnosis and care. Thus, rather than passively recording patient data, the sensor 131 or device 133 takes on an active role in initiating the provisioning of medical care to the patient 10 and starts a cascade of appropriate medical interventions under the tutelage of and followed by physicians and trained healthcare professionals.
In a still further embodiment, the monitor recorder 14 could upload an event detection application to the sensor 131 or device 133 to enable those devices to detect those types of potentially medically-significant events, which would trigger the dispatch of a monitor recorder 14 to the patient 10. Alternatively, the event detection application could be downloaded to the sensor 131 or device 133 from an online application store or similar online application repository. Finally, the monitor recorder 14 could use the sensor 131 or device 133 to generate an appropriate alert, including contacting the patient's physician or healthcare services, via wireless (or wired) communications, upon detecting a potentially medically-significant event or in response to a patient prompting.
The patient 10 could be notified by the sensor 131 or device 133, through the sensor's or device's user interface, that an event of potential medical concern has been detected coupled with an offer to have a monitor recorder 14 sent out to the patient 10, assuming that the patient 10 is not already wearing a monitor recorder 14. Alternatively, the sensor 131 or device 133 could unilaterally send out a request for a monitor recorder 14. The request for a monitor recorder 14 could be sent via wireless (or wired) communications to the patient's physician, a medical service provider organization, a pharmacy, an emergency medical service, or other appropriate healthcare entity that would, in turn, physically provide the patient with a monitor recorder 14. The patient 10 could also be told to pick up a monitor recorder 14 directly from one of the above-identified sources.
Conventional Holter monitors, as well as the ZIO XT Patch and ZIO Event Card devices, described supra, are currently available only by a physician's prescription for a specific patient 10. As a result, the physiological data recorded by these monitors and devices are assumed by healthcare professional to belong to the patient 10. In this prescriptive medicine context, grave questions as to the authenticity of the patient's identity and the data recorded do not generally arise, although current medical practice still favors requesting affirmative patient and caregiver identification at every step of healthcare provisioning. As a device intended for adoption and usage broader than prescriptive medicine, the monitor recorder 14 carries the potential to be used by more than one individual, which can raise concerns as to the veracity of the data recorded.
In a still further embodiment, the mobile device 133, or, if properly equipped, the activity monitor, can be used to help authenticate the patient 10 at the outset of and throughout the monitoring period. The mobile device 133 (or activity monitor) must be appropriately equipped with a digital camera or other feature capable of recording physical indicia located within the proximity of the mobile device 133. For instance, the Samsung Galaxy S5 smartphone has both a biometric fingerprint reader and autofocusing digital camera built in. Upon receipt of a monitor recorder 14, the patient 10 can use the photographic or other recording features of the mobile device 133 (or activity monitor) to physically record the placement and use of the monitor recorder 14. For instance, the patient 10 could take a picture or make a video of the monitor recorder 14 using as applied to the chest using the built-in digital camera. The patient 10 could also swipe a finger over the biometric fingerprint reader. Preferably, the patient 10 would include both his or her face or similar uniquely-identifying marks or indicia, such as a scar, tattoo, body piercing, or RFID chip, plus any visible or electronic indicia on the outside of the monitor recorder's housing, as further described infra with reference to
The mobile device 133 could also serve as a conduit for providing the data collected by the wearable physiology monitor or activity sensor 131 to at least one server 122, or, similarly, the wearable physiology monitor or activity sensor 131 could itself directly provide the collected data to the server 122. The server 122 could then merge the collected data into the wearer's EMRs 134 in the secure database 124, if appropriate (and permissible), or the server 122 could perform an analysis of the collected data, perhaps based by comparison to a population of like wearers of the wearable physiology monitor or activity sensor 131. Still other server 122 functions on the collected data are possible.
Finally, the monitor recorder 14 can also be equipped with a wireless transceiver, as further described infra with reference to
In addition to storing the data received from the monitor 14 recorder as part of the patient's EMR 134, the server 122 can perform other analyses that can speed up the processing of the results of the monitoring. For example, the server 122 executes an adaptive atrial fibrillation detector 160, which divides ECG data collected using the monitor 12 into segments and analyzes each segment to determine whether the patient was experiencing atrial fibrillation at the time the segment was recorded. The adaptive atrial fibrillation detector 160 can be implemented by a convolutional neural network utilizing, for example, a one dimensional formulation for use with ECG data. Additionally, the adaptive atrial fibrillation detector 160 can include hidden layers for performing the classification. In the example described below, two convolutional or pooling hidden layers, and two fully-connected hidden layers are utilized. However, other number of layers are possible.
Prior to being used for classification of the segments, the adaptive atrial fibrillation detector 160 is trained on training data 161, which can be stored in the database 124. The training data 161 includes data files from ambulatory ECG monitors associated with a group of patients. The patients can be selected randomly or identified based on patient condition. In one embodiment, between 200 and 250 files can be collected from different ECG ambulatory monitors, such as monitors 12, though other kinds of monitors are possible, and used as training data. The ambulatory monitors used to collect the ECG data can include the monitor described above with respect to
The database 124 further holds a plurality of ECG features 162, with each feature being one or more elements of an ECG trace, such as particular wave (such as P, Q, R, S, or T waves) or a combination of waves (such as the QRS complex). Waves or combinations of waves of different shapes can be stored as different features 161. For example, a narrow QRS complex and a wide QRS complex can be stored as different features. Similarly, P waves of different height can be stored as different features 162. The features 162 can be extracted from the ECG traces from the training data 161, though other sources are also possible. In one embodiment, a total of 32 features 162 are stored in the database 124, though other numbers of features 162 are possible.
The server 122 receives annotations 163 of patterns of ECG features 162 identified within the training data 161 and provides the annotated patterns 163 to the detector 161. Each annotation states whether a particular pattern of features is associated with atrial fibrillation. For example, a pattern of features can include irregular intervals between two or more successive R waves can be annotated to be associated with atrial fibrillation, though other kinds of patterns are also possible. Not all irregular patterns are necessarily atrial fibrillation. Many other disorders can mimic atrial fibrillation on the basis of just irregular rhythms, such as frequent premature atrial or ventricular contractions or atrial flutter and atrial tachycardia with variable AV conduction. Thus, the P-wave sensitive nature of the Carnation Ambulatory Monitor™ facilitates the machine learning specificity and sensitivity of true atrial fibrillation detection. The annotations 163 are made by a user, such as a physician, or using another classification tool. The annotated patterns 163 of features 162 are run through the detector 160 one or more times, training the detector 160 to recognize patterns of features associated with atrial fibrillation.
While the annotated patterns 163, the ECG features 162, and the training data 161 are shown with reference to
The trained classifier is subsequently used to analyze additional ECG data segments and classify the data as either associated with atrial fibrillation or not, as further described below with reference to
During use, the electrode patch 15 is first adhesed to the skin along the sternal midline 16 (or immediately to either side of the sternum 13). A monitor recorder 14 is then snapped into place on the electrode patch 15 to initiate ECG monitoring.
The electrode patch 15 incorporates features that significantly improve wearability, performance, and patient comfort throughout an extended monitoring period. During wear, the electrode patch 15 is susceptible to pushing, pulling, and torqueing movements, including compressional and torsional forces when the patient bends forward, and tensile and torsional forces when the patient leans backwards. To counter these stress forces, the electrode patch 15 incorporates strain and crimp reliefs, such as described in commonly-assigned U.S. Pat. No. 9,545,204, issued Jan. 17, 2017, the disclosure of which is incorporated by reference. In addition, the cut-outs 22 and longitudinal midsection 23 help minimize interference with and discomfort to breast tissue, particularly in women (and gynecomastic men). The cut-outs 22 and longitudinal midsection 23 further allow better conformity of the electrode patch 15 to sternal bowing and to the narrow isthmus of flat skin that can occur along the bottom of the intermammary cleft between the breasts, especially in buxom women. The cut-outs 22 and longitudinal midsection 23 help the electrode patch 15 fit nicely between a pair of female breasts in the intermammary cleft. Still other shapes, cut-outs and conformities to the electrode patch 15 are possible.
The monitor recorder 14 removably and reusably snaps into an electrically non-conductive receptacle 25 during use. The monitor recorder 14 contains electronic circuitry for recording and storing the patient's electrocardiography as sensed via a pair of ECG electrodes provided on the electrode patch 15, such as described in commonly-assigned U.S. Pat. No. 9,730,593, issued Aug. 15, 2017, the disclosure which is incorporated by reference. The non-conductive receptacle 25 is provided on the top surface of the flexible backing 20 with a retention catch 26 and tension clip 27 molded into the non-conductive receptacle 25 to conformably receive and securely hold the monitor recorder 14 in place.
The monitor recorder 14 includes a sealed housing that snaps into place in the non-conductive receptacle 25.
The electrode patch 15 is intended to be disposable. The monitor recorder 14, however, is reusable and can be transferred to successive electrode patches 15 to ensure continuity of monitoring. The placement of the wearable monitor 12 in a location at the sternal midline 16 (or immediately to either side of the sternum 13) benefits long-term extended wear by removing the requirement that ECG electrodes be continually placed in the same spots on the skin throughout the monitoring period. Instead, the patient is free to place an electrode patch 15 anywhere within the general region of the sternum 13.
As a result, at any point during ECG monitoring, the patient's skin is able to recover from the wearing of an electrode patch 15, which increases patient comfort and satisfaction, while the monitor recorder 14 ensures ECG monitoring continuity with minimal effort. A monitor recorder 14 is merely unsnapped from a worn out electrode patch 15, the worn out electrode patch 15 is removed from the skin, a new electrode patch 15 is adhered to the skin, possibly in a new spot immediately adjacent to the earlier location, and the same monitor recorder 14 is snapped into the new electrode patch 15 to reinitiate and continue the ECG monitoring.
During use, the electrode patch 15 is first adhered to the skin in the sternal region.
In addition, a battery compartment 36 is formed on the bottom surface of the non-conductive receptacle 25, and a pair of battery leads (not shown) electrically interface the battery to another pair of the electrical pads 34. The battery contained within the battery compartment 35 can be replaceable, rechargeable or disposable.
The monitor recorder 14 draws power externally from the battery provided in the non-conductive receptacle 25, thereby uniquely obviating the need for the monitor recorder 14 to carry a dedicated power source.
The placement of the flexible backing 20 on the sternal midline 16 (or immediately to either side of the sternum 13) also helps to minimize the side-to-side movement of the wearable monitor 12 in the left- and right-handed directions during wear. To counter the dislodgment of the flexible backing 20 due to compressional and torsional forces, a layer of non-irritating adhesive, such as hydrocolloid, is provided at least partially on the underside, or contact, surface of the flexible backing 20, but only on the distal end 30 and the proximal end 31. As a result, the underside, or contact surface of the longitudinal midsection 23 does not have an adhesive layer and remains free to move relative to the skin. Thus, the longitudinal midsection 23 forms a crimp relief that respectively facilitates compression and twisting of the flexible backing 20 in response to compressional and torsional forces. Other forms of flexible backing crimp reliefs are possible.
Unlike the flexible backing 20, the flexible circuit 32 is only able to bend and cannot stretch in a planar direction. The flexible circuit 32 can be provided either above or below the flexible backing 20.
ECG monitoring and other functions performed by the monitor recorder 14 are provided through a micro controlled architecture.
Operation of the circuitry 60 of the monitor recorder 14 is managed by a microcontroller 61. The micro-controller 61 includes a program memory unit containing internal flash memory that is readable and writeable. The internal flash memory can also be programmed externally. The micro-controller 61 draws power externally from the battery provided on the electrode patch 15 via a pair of the electrical contacts 56. The microcontroller 61 connects to the ECG front end circuit 63 that measures raw cutaneous electrical signals and generates an analog ECG signal representative of the electrical activity of the patient's heart over time.
The circuitry 60 of the monitor recorder 14 also includes a flash memory 62, which the micro-controller 61 uses for storing ECG monitoring data and other physiology and information. The flash memory 62 also draws power externally from the battery provided on the electrode patch 15 via a pair of the electrical contacts 56. Data is stored in a serial flash memory circuit, which supports read, erase and program operations over a communications bus. The flash memory 62 enables the microcontroller 61 to store digitized ECG data. The communications bus further enables the flash memory 62 to be directly accessed externally over the external connector 65 when the monitor recorder 14 is interfaced to a download station.
The circuitry 60 of the monitor recorder 14 further includes an actigraphy sensor 64 implemented as a 3-axis accelerometer. The accelerometer may be configured to generate interrupt signals to the microcontroller 61 by independent initial wake up and free fall events, as well as by device position. In addition, the actigraphy provided by the accelerometer can be used during post-monitoring analysis to correct the orientation of the monitor recorder 14 if, for instance, the monitor recorder 14 has been inadvertently installed upside down, that is, with the monitor recorder 14 oriented on the electrode patch 15 towards the patient's feet, as well as for other event occurrence analyses, such as described in commonly-assigned U.S. Pat. No. 9,737,224, issued Aug. 22, 2017, the disclosure of which is incorporated by reference.
The circuitry 60 of the monitor recorder 14 includes a wireless transceiver 69 that can provides wireless interfacing capabilities. The wireless transceiver 69 also draws power externally from the battery provided on the electrode patch 15 via a pair of the electrical contacts 56. The wireless transceiver 69 can be implemented using one or more forms of wireless communications, including the IEEE 802.11 computer communications standard, that is Wi-Fi; the 4G mobile phone mobile standard; the Bluetooth data exchange standard; or other wireless communications or data exchange standards and protocols. The type of wireless interfacing capability could limit the range of interoperability of the monitor recorder 14; for instance, Bluetooth-based implementations are designed for low power consumption with a short communications range.
The microcontroller 61 includes an expansion port that also utilizes the communications bus. External devices, separately drawing power externally from the battery provided on the electrode patch 15 or other source, can interface to the microcontroller 61 over the expansion port in half duplex mode. For instance, an external physiology sensor can be provided as part of the circuitry 60 of the monitor recorder 14, or can be provided on the electrode patch 15 with communication with the micro-controller 61 provided over one of the electrical contacts 56. The physiology sensor can include an SpO2 sensor, blood pressure sensor, temperature sensor, respiratory rate sensor, glucose sensor, airflow sensor, volumetric pressure sensing, or other types of sensor or telemetric input sources. For instance, the integration of an airflow sensor is described in commonly-assigned U.S. Pat. No. 9,364,155, issued Jun. 14, 2016, the disclosure which is incorporated by reference.
Finally, the circuitry 60 of the monitor recorder 14 includes patient-interfaceable components, including a tactile feedback button 66, which a patient can press to mark events or to perform other functions, and a buzzer 67, such as a speaker, magnetic resonator or piezoelectric buzzer. The buzzer 67 can be used by the microcontroller 61 to output feedback to a patient such as to confirm power up and initiation of ECG monitoring. Still other components as part of the circuitry 60 of the monitor recorder 14 are possible.
While the monitor recorder 14 operates under micro control, most of the electrical components of the electrode patch 15 operate passively.
The circuitry 70 of the electrode patch 15 performs three primary functions. First, a battery 71 is provided in a battery compartment formed on the bottom surface of the non-conductive receptacle 25. The battery 71 is electrically interfaced to the circuitry 60 of the monitor recorder 14 as a source of external power. The unique provisioning of the battery 71 on the electrode patch 15 provides several advantages. First, the locating of the battery 71 physically on the electrode patch 15 lowers the center of gravity of the overall wearable monitor 12 and thereby helps to minimize shear forces and the effects of movements of the patient and clothing. Moreover, the housing 50 of the monitor recorder 14 is sealed against moisture and providing power externally avoids having to either periodically open the housing 50 for the battery replacement, which also creates the potential for moisture intrusion and human error, or to recharge the battery, which can potentially take the monitor recorder 14 off line for hours at a time. In addition, the electrode patch 15 is intended to be disposable, while the monitor recorder 14 is a reusable component. Each time that the electrode patch 15 is replaced, a fresh battery is provided for the use of the monitor recorder 14, which enhances ECG monitoring performance quality and duration of use. Finally, the architecture of the monitor recorder 14 is open, in that other physiology sensors or components can be added by virtue of the expansion port of the microcontroller 61. Requiring those additional sensors or components to draw power from a source external to the monitor recorder 14 keeps power considerations independent of the monitor recorder 14. Thus, a battery of higher capacity could be introduced when needed to support the additional sensors or components without effecting the monitor recorders circuitry 60.
Second, the pair of ECG electrodes 38, 39 respectively provided on the distal and proximal ends of the flexible circuit 32 are electrically coupled to the set of pads 34 provided on the bottom of the non-conductive receptacle 25 by way of their respective circuit traces 33, 37. The signal ECG electrode 39 includes a protection circuit 72, which is an inline resistor that protects the patient from excessive leakage current.
Last, in a further embodiment, the circuitry 70 of the electrode patch 15 includes a cryptographic circuit 73 to authenticate an electrode patch 15 for use with a monitor recorder 14. The cryptographic circuit 73 includes a device capable of secure authentication and validation. The cryptographic device 73 ensures that only genuine, non-expired, safe, and authenticated electrode patches 15 are permitted to provide monitoring data to a monitor recorder 14, such as described in commonly-assigned U.S. Pat. No. 9,655,538, issued May 23, 2017, the disclosure which is incorporated by reference.
In a further embodiment, the circuitry 70 of the electrode patch 15 includes a wireless transceiver 75, in lieu the including of the wireless transceiver 69 in the circuitry 60 of the monitor recorder 14, which interfaces with the microcontroller 61 over the microcontroller's expansion port via the external connector 74.
The monitor recorder 14 continuously monitors the patient's heart rate and physiology.
Following satisfactory completion of the power up sequence, an iterative processing loop (steps 102-109) is continually executed by the microcontroller 61. During each iteration (step 102) of the processing loop, the ECG frontend 63 (shown in
Sampling of the R-to-R interval enables heart rate information derivation. For instance, the R-to-R interval represents the ventricular rate and rhythm, while the P-to-P interval represents the atrial rate and rhythm. Importantly, the PR interval is indicative of atrioventricular (AV) conduction time and abnormalities in the PR interval can reveal underlying heart disorders, thus representing another reason why the P-wave quality achievable by the extended wear ambulatory electrocardiography and physiological sensor monitor described herein is medically unique and important. The long-term observation of these ECG indicia, as provided through extended wear of the wearable monitor 12, provides valuable insights to the patient's cardiac function and overall well-being.
Each sampled ECG signal, in quantized and digitized form, is temporarily staged in buffer (step 105), pending compression preparatory to storage in the flash memory 62 (step 106). Following compression, the compressed ECG digitized sample is again buffered (step 107), then written to the flash memory 62 (step 108) using the communications bus. Processing continues (step 109), so long as the monitoring recorder 14 remains connected to the electrode patch 15 (and storage space remains available in the flash memory 62), after which the processing loop is exited and execution terminates. Still other operations and steps are possible.
In a further embodiment, the monitor recorder 14 also continuously receives data from wearable physiology monitors or activity sensors 131 and mobile devices 133 (shown in
The monitor recorder 14 stores ECG data and other information in the flash memory 62 (shown in
Initially, the download station 125 is connected to the monitor recorder 14 (step 151), such as by physically interfacing to a set of terminals 128 on a paired receptacle 127 or by wireless connection, if available. The data stored on the monitor recorder 14, including ECG and physiological monitoring data, other recorded data, and other information are retrieved (step 152) over a hard link 135 using a control program 137 (“Ctl”) or analogous application executing on a personal computer 136 or other connectable computing device.
The data retrieved from the monitor recorder 14 is in a proprietary storage format and each datum of recorded ECG monitoring data, as well as any other physiological data or other information, must be converted, so that the data can be used by a third-party post-monitoring analysis program. Each datum of ECG monitoring data is converted by the middleware (steps 153-159) in an iterative processing loop. During each iteration (step 153), the ECG datum is read (step 154) and, if necessary, the gain of the ECG signal is adjusted (step 155) to compensate, for instance, for relocation or replacement of the electrode patch 15 during the monitoring period.
In addition, depending upon the configuration of the wearable monitor 12, other physiological data (or other information), including patient events, such as a fall, peak activity level, sleep detection, Detection of patient activity levels and states, and so on, may be recorded along with the ECG monitoring data. For instance, actigraphy data may have been sampled by the actigraphy sensor 64 based on a sensed event occurrence, such as a sudden change in orientation due to the patient taking a fall. In response, the monitor recorder 14 will embed the actigraphy data samples into the stream of data, including ECG monitoring data, that is recorded to the flash memory 62 by the micro-controller 61. Post-monitoring, the actigraphy data is temporally matched to the ECG data to provide the proper physiological context to the sensed event occurrence. As a result, the three-axis actigraphy signal is turned into an actionable event occurrence that is provided, through conversion by the middleware, to third party post-monitoring analysis programs, along with the ECG recordings contemporaneous to the event occurrence. Other types of processing of the other physiological data (or other information) are possible.
Thus, during execution of the middleware, any other physiological data (or other information) that has been embedded into the recorded ECG monitoring data is read (step 156) and time-correlated to the time frame of the ECG signals that occurred at the time that the other physiological data (or other information) was noted (step 157). Finally, the ECG datum, signal gain adjusted, if appropriate, and other physiological data, if applicable and as time-correlated, are stored in a format suitable to the backend software (step 158) used in post-monitoring analysis.
In a further embodiment, the other physiological data, if apropos, is embedded within an unused ECG track. For example, the SCP-ENG standard allows multiple ECG channels to be recorded into a single ECG record. The monitor recorder 14, though, only senses one ECG channel. The other physiological data can be stored into an additional ECG channel, which would otherwise be zero-padded or altogether omitted. The backend software would then be able to read the other physiological data in context with the single channel of ECG monitoring data recorded by the monitor recorder 14, provided the backend software implemented changes necessary to interpret the other physiological data. Still other forms of embedding of the other physiological data with formatted ECG monitoring data, or of providing the other physiological data in a separate manner, are possible.
Processing continues (step 159) for each remaining ECG datum, after which the processing loop is exited and execution terminates. Still other operations and steps are possible.
Using machine-learning to detect atrial fibrillation using patterns of ECG features that are known to be indicative of atrial fibrillation allows more flexibility in detecting instances of atrial fibrillation than when the detection is based on a single set of predefined parameters.
ECG data collected from a patient via a ECG monitor, such as the monitor described above with respect to
Further, a detected pattern can be added to the ECG training data and subsequently used to train the adaptive atrial fibrillation detector. Other actions are also possible.
Accurate detection of presence of atrial fibrillation during ECG monitoring of the ECG data segments is dependent on training accuracy.
The training data is divided into segments, with each segment corresponding to a temporal window during which the portion of the ECG signal in that segment was recorded (step 182). In one embodiment, the temporal window can be between 2 and 12 seconds; for example in one embodiment, the temporal window can be 8.5 seconds. Patterns of ECG features within the segments are annotated as being indicative of atrial fibrillation or not being indicative of atrial fibrillation (step 183). In one embodiment, a total of 32 unique patterns of ECG features is annotated, though other numbers of combinations are also possible.
The annotated patterns are then provided to the adaptive atrial fibrillation detector (step 184). In one embodiment, the patterns are each provided and run through the adaptive atrial fibrillation detector once. In a further embodiment, the patterns are run through the adaptive atrial fibrillation detector multiple times.
After the training data has been run through the adaptive atrial fibrillation detector, testing can be performed to determine classification accuracy of the adaptive atrial fibrillation detector (step 185). During testing, a set of testing data, such as ECG monitoring results collected by ambulatory ECG monitors, such as described above with reference to
Once the adaptive atrial fibrillation detector is accurately trained, further ECG data is collected and provided to the detector for classification.
During the first convolution layer, ECG trace features are identified (block 193) using, for example, sliding filters, though other identification techniques are possible. Examples of ECG trace features can include R waves without clearly discernible P waves, as well as other types of features that are indicative of atrial fibrillation. For example, other patterns indicative of AF can further include highly disorganized P-waves that do not settle into a repeating pattern. In one embodiment, filters for at least 32 features are run against the ECG data. During the second convolution layer, repeating patterns of the features are identified (step 194), including, for example, irregular R intervals between successive R-waves, though other patterns are possible.
Next, the data obtained from the second convolution layer is provided to a first fully connected cross-connection layer, which builds (step 195) a matrix with the repeating features representing the columns and matrix multipliers representing rows. An intersection value for each combination of the repeating features and matrix multipliers are listed in the matrix as cross connection weights. Specifically, the intersection value can be determined by multiplying each repeating feature value with a matrix multiplier and scaling the product by a predetermined factor. However, other methods for determining the intersection values are possible.
The second fully connected cross-connection layer utilizes the cross-connection weights from the first fully connected cross-connection layer and multiplies the cross-connection weights by further weights to calculate (step 196) final cross-connection values for each ECG data segment. The final cross-connection values include an atrial fibrillation classification value, which is indicative of a probability that the patient experienced atrial fibrillation during the temporal interval during which the ECG data segment was recorded, and a non-atrial fibrillation data classification value, which is indicative of the probability that the patient did not experience atrial fibrillation during the temporal period during which the ECG data segment was recorded.
Based on the final cross-connection values, a determination is made as to whether the atrial fibrillation classification value for each data segment exceeds the non-atrial fibrillation classification value (step 197). The ECG data segments for which the atrial fibrillation classification value does exceed the non-atrial fibrillation classification value are classified as associated with atrial fibrillation of the patient (step 198), signifying that the patient experienced atrial fibrillation during at least a portion of the temporal windows during which these ECG data segments were recorded. For those ECG data segments for which the atrial fibrillation classification value does not exceed the non-atrial fibrillation classification value (197), a non-atrial fibrillation classification is assigned (step 199), signifying that the patient did not experience atrial fibrillation during the temporal windows during which these ECG data segments were recorded.
While the invention has been particularly shown and described as referenced to the embodiments thereof, those skilled in the art will understand that the foregoing and other changes in form and detail may be made therein without departing from the spirit and scope.
Bardy, Gust H., Dreisbach, Ezra M., Boleyn, Rodney, Dulken, Chuck
| Patent | Priority | Assignee | Title |
| 10813565, | Oct 31 2014 | iRhythm Technologies, Inc. | Wearable monitor |
| 11083371, | Feb 12 2020 | iRhythm Technologies, Inc | Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless network |
| 11141091, | May 12 2010 | iRhythm Technologies, Inc. | Device features and design elements for long-term adhesion |
| 11246523, | Aug 06 2020 | iRhythm Technologies, Inc | Wearable device with conductive traces and insulator |
| 11246524, | Feb 12 2020 | iRhythm Technologies, Inc | Non-invasive cardiac monitor and methods of using recorded cardiac data to infer a physiological characteristic of a patient |
| 11253185, | Feb 12 2020 | iRhythm Technologies, Inc. | Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless network |
| 11253186, | Feb 12 2020 | iRhythm Technologies, Inc. | Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless network |
| 11289197, | Oct 31 2014 | iRhythm Technologies, Inc. | Wearable monitor |
| 11337632, | Aug 06 2020 | iRhythm Technologies, Inc | Electrical components for physiological monitoring device |
| 11350864, | Aug 06 2020 | iRhythm Technologies, Inc | Adhesive physiological monitoring device |
| 11350865, | Aug 06 2020 | iRhythm Technologies, Inc | Wearable device with bridge portion |
| 11355244, | May 06 2019 | Medtronic, Inc | Machine learning based depolarization identification and arrhythmia localization visualization |
| 11375941, | Feb 12 2020 | iRhythm Technologies, Inc. | Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless network |
| 11382555, | Feb 12 2020 | iRhythm Technologies, Inc. | Non-invasive cardiac monitor and methods of using recorded cardiac data to infer a physiological characteristic of a patient |
| 11399760, | Aug 06 2020 | iRhythm Technologies, Inc. | Wearable device with conductive traces and insulator |
| 11443852, | May 06 2019 | Medtronic, Inc. | Reduced power machine learning system for arrhythmia detection |
| 11475998, | May 06 2019 | Medtronic, Inc. | Data preparation for artificial intelligence-based cardiac arrhythmia detection |
| 11497432, | Feb 12 2020 | iRhythm Technologies, Inc. | Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless |
| 11504041, | Aug 06 2020 | iRhythm Technologies, Inc. | Electrical components for physiological monitoring device |
| 11583687, | May 06 2019 | Medtronic, Inc.; Medtronic, Inc | Selection of probability thresholds for generating cardiac arrhythmia notifications |
| 11589792, | Aug 06 2020 | iRhythm Technologies, Inc. | Wearable device with bridge portion |
| 11596344, | Oct 30 2019 | TENCENT AMERICA LLC | Deep neural network on ECG Poincare plot for atrial fibrillation classification |
| 11605458, | Oct 31 2014 | iRhythm Technologies, Inc | Wearable monitor |
| 11617533, | May 06 2019 | Medtronic, Inc. | Visualization of arrhythmia detection by machine learning |
| 11627902, | Jan 24 2013 | iRhythm Technologies, Inc. | Physiological monitoring device |
| 11679268, | Oct 05 2018 | Medtronic, Inc. | Multi-tier prediction of cardiac tachyarrythmia |
| 11694804, | May 06 2019 | Medtronic, Inc. | Reduced power machine learning system for arrhythmia detection |
| 11696718, | May 06 2019 | Medtronic, Inc. | Arrhythmia detection with feature delineation and machine learning |
| 11723577, | May 06 2019 | Medtronic, Inc.; Medtronic, Inc | Visualization of arrhythmia detection by machine learning |
| 11751789, | Aug 06 2020 | iRhythm Technologies, Inc. | Wearable device with conductive traces and insulator |
| 11756684, | Oct 31 2014 | iRhythm Technologies, Inc. | Wearable monitor |
| 11776691, | May 06 2019 | Medtronic, Inc. | Machine learning based depolarization identification and arrhythmia localization visualization |
| 11806150, | Aug 06 2020 | iRhythm Technologies, Inc. | Wearable device with bridge portion |
| 11925469, | Feb 12 2020 | iRhythm Technologies, Inc. | Non-invasive cardiac monitor and methods of using recorded cardiac data to infer a physiological characteristic of a patient |
| Patent | Priority | Assignee | Title |
| 10034614, | Feb 29 2012 | General Electric Company | Fractional flow reserve estimation |
| 10045708, | Aug 26 2013 | Apple Inc. | Method of detecting the wearing limb of a wearable electronic device |
| 10049182, | Feb 03 2011 | ISANSYS LIFECARE LIMITED | Health monitoring |
| 3215136, | |||
| 3569852, | |||
| 3602215, | |||
| 3699948, | |||
| 3718772, | |||
| 3893453, | |||
| 4123785, | Aug 25 1976 | Del Mar Avionics | Recorder for cardiac signals with manually actuated event marking |
| 4151513, | Mar 06 1975 | Medtronic, Inc. | Apparatus for sensing and transmitting a pacemaker's stimulating pulse |
| 4328814, | Jun 04 1980 | The Kendall Company | Precordial ECG strip |
| 4441500, | Apr 17 1980 | FERRIS PHARMACEUTICALS INC | EKG Electrode |
| 4532934, | Nov 01 1978 | Del Mar Avionics | Pacemaker monitoring recorder and malfunction analyzer |
| 4546342, | Dec 14 1983 | Digital Recording Research Limited Partnership | Data compression method and apparatus |
| 4550502, | Apr 15 1983 | Device for analysis of recorded electrocardiogram | |
| 4580572, | Jun 01 1983 | Bio-Stimu Trend Corporation | Garment apparatus for delivering or receiving electric impulses |
| 4635646, | Nov 10 1983 | Portable apparatus for monitoring heart activity | |
| 4653022, | Aug 09 1984 | Kabushiki Kaisha Tatebe Seishudo | Portable electrocardiogram storing apparatus |
| 4716903, | Oct 06 1986 | Pacesetter, Inc | Storage in a pacemaker memory |
| 4809705, | Oct 11 1984 | Portable electrocardiogram monitor | |
| 4915656, | Oct 21 1988 | PHYSIO-CONTROL, INC | Discriminating medical electrode connector |
| 5007429, | Aug 21 1987 | PULSETREND, INC , A CORP OF DE | Interface using 12-digit keypad for programming parameters in ambulatory blood pressure monitor |
| 5025794, | Aug 30 1988 | ARRHYTHMIA RESEARCH TECHNOLOGY, INC | Method for analysis of electrocardiographic signal QRS complex |
| 5107480, | Oct 23 1989 | U.S. Philips Corporation | Device for reading digital information recorded on an information carrier, and peak detector and information carrier for such a device |
| 5168876, | Oct 30 1990 | AHP LEASING, INC ; COROMETRICS MEDICAL SYSTEMS, INC ; AHP SUBSIDIARY HOLDING CORPORATION | Connector device for use in monitoring fetal heart rate |
| 5215098, | Aug 12 1991 | Pacesetter, Inc | Data compression of cardiac electrical signals using scanning correlation and temporal data compression |
| 5231990, | Jul 09 1992 | Spacelabs Healthcare, LLC | Application specific integrated circuit for physiological monitoring |
| 5263481, | May 21 1992 | Electrode system with disposable gel | |
| 5265579, | Sep 21 1992 | X-ray transparent monitoring electrode and method for making | |
| 5333615, | Jun 22 1992 | Apparatus for digitally recording and analyzing electrocardial and other bioelectric signals | |
| 5341806, | Apr 18 1991 | PHYSIO-CONTROL CORPORATION A CORP OF DELAWARE | Multiple electrode strip |
| 5348008, | Nov 25 1991 | Oishi-Kogyo Company | Cardiorespiratory alert system |
| 5355891, | Jan 13 1992 | Huntleigh Technology Limited | ECG analyzer |
| 5365934, | Jun 28 1991 | Brunswick Corporation | Apparatus and method for measuring heart rate |
| 5365935, | Sep 10 1991 | CARD GUARD TECHNOLOGIES, INC | Portable, multi-channel ECG data monitor/recorder |
| 5392784, | Aug 20 1993 | Koninklijke Philips Electronics N V | Virtual right leg drive and augmented right leg drive circuits for common mode voltage reduction in ECG and EEG measurements |
| 5402780, | Sep 02 1993 | Medical electrode with offset contact stud | |
| 5402884, | Sep 24 1992 | Cardiac Science Corporation | Medical electrode packaging technology |
| 5450845, | Jan 11 1993 | Medical electrode system | |
| 5451876, | Oct 18 1993 | General Electric Company | MRI system with dynamic receiver gain |
| 5458141, | Aug 04 1993 | MORTARA INSTRUMENT, INC | Abrasive skin electrode |
| 5473537, | Jul 30 1993 | BEHAVIORAL ALGORITHMS, INC | Method for evaluating and reviewing a patient's condition |
| 5483969, | Sep 21 1994 | Medtronic, Inc. | Method and apparatus for providing a respiratory effort waveform for the treatment of obstructive sleep apnea |
| 5511553, | Feb 15 1989 | Device-system and method for monitoring multiple physiological parameters (MMPP) continuously and simultaneously | |
| 5540733, | Sep 21 1994 | INSPIRE MEDICAL SYSTEMS, INC | Method and apparatus for detecting and treating obstructive sleep apnea |
| 5546952, | Sep 21 1994 | Medtronic, Inc. | Method and apparatus for detection of a respiratory waveform |
| 5549655, | Sep 21 1994 | INSPIRE MEDICAL SYSTEMS, INC | Method and apparatus for synchronized treatment of obstructive sleep apnea |
| 5579919, | Sep 24 1992 | Cardiac Science Corporation | Medical electrode packaging technology |
| 5582181, | Sep 10 1994 | Koninklijke Philips Electronics N V | Device and a method for equalizing a patient's potential with respect to medical equipment |
| 5601089, | Mar 12 1993 | Koninklijke Philips Electronics N V | Method and apparatus for boosting the amplitude of ECG signals within a predetermined frequency range |
| 5623935, | Jul 07 1994 | ELA Medical, S.A. | Data compression methods and apparatus for use with physiological data |
| 5682901, | Aug 03 1993 | Method and apparatus for measuring autonomic activity of a patient | |
| 5697955, | May 10 1996 | Cardiac Science Corporation | Defibrillator electrodes and date code detector circuit |
| 5724967, | Nov 21 1995 | Nellcor Puritan Bennett Incorporated | Noise reduction apparatus for low level analog signals |
| 5749902, | May 22 1996 | Cardiac Science Corporation | Recorded data correction method and apparatus for isolated clock systems |
| 5788633, | Jan 28 1997 | Koninklijke Philips Electronics N V | ECG electrode strip with elongated slots |
| 5817151, | Jun 04 1996 | ZOLL Medical Corporation | Circuit detectable packaged medical electrodes |
| 5819741, | Oct 07 1994 | Ortivus AB | Cardiac monitoring system and method |
| 5850920, | Sep 24 1992 | ZOLL Medical Corporation | Medical electrode packaging technology |
| 5876351, | Apr 10 1997 | Mitchell Rohde | Portable modular diagnostic medical device |
| 5906583, | Aug 20 1997 | R.Z. Comparative Diagnostics Ltd. | Automatic cardiometer |
| 5951598, | Jan 14 1997 | Koninklijke Philips Electronics N V | Electrode system |
| 5957857, | May 07 1998 | Cardiac Pacemakers, Inc. | Apparatus and method for automatic sensing threshold determination in cardiac pacemakers |
| 5984102, | Sep 24 1992 | Cardiac Science Corporation | Medical electrode packaging technology |
| 6032064, | Oct 11 1996 | Nellcor Puritan Bennett LLC | Electrode array system for measuring electrophysiological signals |
| 6038469, | Oct 07 1994 | Ortivus AB | Myocardial ischemia and infarction analysis and monitoring method and apparatus |
| 6101413, | Jun 04 1996 | ZOLL Medical Corporation | Circuit detectable pediatric defibrillation electrodes |
| 6115638, | May 04 1998 | ZOLL Medical Corporation | Medical electrode with conductive release liner |
| 6117077, | Jan 22 1999 | Spacelabs Healthcare, LLC | Long-term, ambulatory physiological recorder |
| 6134479, | Jul 09 1998 | ZOLL Medical Corporation | Electrode triad for external defibrillation |
| 6148233, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillation system having segmented electrodes |
| 6149602, | Sep 10 1996 | User-worn electrocardiogram viewer device | |
| 6149781, | Jan 10 1994 | ELECTROPLATING TECHNOLOGIES, LTD | Method and apparatus for electrochemical processing |
| 6188407, | Mar 04 1998 | GE MEDICAL SYSTEMS INFORMATION TECHNOLOGIES, INC | Reconfigurable user interface for modular patient monitor |
| 6245025, | Aug 10 1995 | REMOTAID HUNGARY KFT | Method and apparatus for measuring fetal heart rate and an electroacoustic sensor for receiving fetal heart sounds |
| 6246330, | May 29 1998 | Elimination-absorber monitoring system | |
| 6249696, | Jan 15 1999 | PHYSIO-CONTROL, INC | Method and apparatus for increasing the low frequency dynamic range of a digital ECG measuring system |
| 6269267, | Sep 02 1998 | Koninklijke Philips Electronics N V | Configurable arrhythmia analysis algorithm with security interface |
| 6272385, | Sep 01 1998 | Koninklijke Philips Electronics N V | Independently deployable sealed defibrillator electrode pad and method of use |
| 6298255, | Jun 09 1999 | Covidien LP | Smart electrophysiological sensor system with automatic authentication and validation and an interface for a smart electrophysiological sensor system |
| 6301502, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillation system |
| 6304773, | May 21 1998 | PHYSIO-CONTROL, INC | Automatic detection and reporting of cardiac asystole |
| 6304780, | Mar 07 1997 | ZOLL Medical Corporation | External defibrillator system with diagnostic module |
| 6304783, | Oct 14 1999 | Philips Electronics North America Corporation | Defibrillator system including a removable monitoring electrodes adapter and method of detecting the monitoring adapter |
| 6374138, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillation system |
| 6381482, | May 13 1998 | Georgia Tech Research Corporation | Fabric or garment with integrated flexible information infrastructure |
| 6416471, | Apr 15 1999 | CLEARPATH PARTNERS, LLC | Portable remote patient telemonitoring system |
| 6418342, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillation system |
| 6424860, | Oct 07 1994 | Ortivus AB | Myocardial ischemia and infarction analysis and monitoring method and apparatus |
| 6427083, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillation system |
| 6427085, | May 11 1999 | Cardiac Pacemakers, Inc.; Cardiac Pacemakers, Inc | Cardiac sense amplifier for capture verification |
| 6454708, | Apr 15 1999 | CLEARPATH PARTNERS, LLC | Portable remote patient telemonitoring system using a memory card or smart card |
| 6456872, | Jun 23 1999 | Sorin CRM SAS | Holter-type apparatus for recording physiological signals indicative of cardiac activity |
| 6463320, | Dec 22 1999 | GE MEDICAL SYSTEMS INFORMATION TECHNOLOGIES, INC | Clinical research workstation |
| 6546285, | Mar 07 1997 | ZOLL Medical Corporation | Long term wear electrode for defibrillation system |
| 6605046, | Jun 03 1991 | Spacelabs Healthcare, LLC | Ambulatory physio-kinetic monitor with envelope enclosure |
| 6607485, | Jun 03 1999 | Cardiac Intelligence Corporation | Computer readable storage medium containing code for automated collection and analysis of patient information retrieved from an implantable medical device for remote patient care |
| 6611705, | Jul 17 2001 | Lifesync Corporation | Wireless electrocardiograph system and method |
| 6671545, | Mar 07 1997 | ZOLL Medical Corporation | Defibrillator with controller operating in a low power mode |
| 6671547, | Jun 13 2001 | Koninklijke Philips Electronics N V | Adaptive analysis method for an electrotherapy device and apparatus |
| 6694186, | Nov 16 1999 | Cardiac Intelligence Corporation | Automated collection and analysis patient care system for managing the pathophysiological outcomes of atrial fibrillation |
| 6704595, | Nov 16 1999 | Cardiac Intelligence Corporation | Automated method for diagnosing and monitoring the outcomes of atrial fibrillation |
| 6705991, | Jun 03 1999 | Cardiac Intelligence Corporation | System and method for providing patient status diagnosis for use in automated patient care |
| 6719701, | Jan 28 2002 | Pacesetter, Inc. | Implantable syncope monitor and method of using the same |
| 6754523, | Nov 28 2000 | Method of analysis of the electrocardiogram | |
| 6782293, | Sep 14 2001 | ZOLL Medical Corporation | Defibrillation electrode assembly including CPR pad |
| 6856832, | Dec 25 1997 | NIHON KOHDEN CORPORATION | Biological signal detection apparatus Holter electrocardiograph and communication system of biological signals |
| 6860897, | Jun 03 1999 | Cardiac Intelligence Corporation | System and method for providing feedback to an individual patient for automated remote patient care |
| 6866629, | Jul 26 1999 | Cardiac Intelligence Corporation | Automated system and method for establishing a patient status reference baseline |
| 6887201, | Jul 26 1999 | Cardiac Intelligence Corporation | System and method for determining a reference baseline of regularly retrieved patient information for automated remote patient care |
| 6893397, | Jul 26 1999 | Cardiac Intelligence Corporation | System and method for determining a reference baseline of patient information for automated remote patient care |
| 6895261, | Jul 13 2000 | Portable, wireless communication apparatus integrated with garment | |
| 6904312, | Nov 16 1999 | Cardiac Intelligence Corporation | System and method for diagnosing and monitoring outcomes of atrial fibrillation for automated remote patient care |
| 6908431, | Jun 03 1999 | Cardiac Intelligence Corporation | System and method for providing feedback to an individual patient for automated remote patient care |
| 6913577, | Nov 16 1999 | Cardiac Intelligence Corporation | System and method for diagnosing and monitoring myocardial ischemia for automated remote patient care |
| 6944498, | Mar 07 1997 | ZOLL Medical Corporation | Method of utilizing an external defibrillator by replacing its electrodes |
| 6960167, | Jul 26 1999 | Cardiac Intelligence Corporation | System and method for patient monitoring using a reference baseline for use in automated patient care |
| 6970731, | Sep 21 1998 | Georgia Tech Research Corp.; Georgia Tech Research Corporation | Fabric-based sensor for monitoring vital signs |
| 6978169, | Apr 04 2002 | BIOMEDICA DIAGNOSTICS INC | Personal physiograph |
| 6993377, | Feb 22 2002 | The Board of Trustees of the University of Arkansas | Method for diagnosing heart disease, predicting sudden death, and analyzing treatment response using multifractal analysis |
| 7020508, | Aug 22 2002 | JB IP ACQUISITION LLC | Apparatus for detecting human physiological and contextual information |
| 7027864, | Apr 17 2002 | Koninklijke Philips Electronics N.V. | Defibrillation system and method designed for rapid attachment |
| 7065401, | May 08 2002 | ZOLL Medical Corporation | Method of applying electrical signals to a patient and automatic wearable external defibrillator |
| 7085601, | Nov 17 1999 | Koninklijke Philips Electronics N V | External atrial defibrillator and method for personal termination of atrial fibrillation |
| 7104955, | Jun 03 1999 | Cardiac Pacemakers, Inc | System and method for collection and analysis of regularly retrieved patient information for automated remote patient care |
| 7134996, | Jun 03 1999 | Cardiac Pacemakers, Inc | System and method for collection and analysis of patient information for automated remote patient care |
| 7137389, | Sep 23 1996 | ResMed Limited | Method and apparatus for determining instantaneous inspired volume of a subject during ventilatory assistance |
| 7147600, | Jul 26 1999 | Cardiac Intelligence Corporation | System and method for determining a reference baseline of patient information |
| 7215991, | Sep 04 1993 | Body Science LLC | Wireless medical diagnosis and monitoring equipment |
| 7248916, | Jul 26 1999 | Cardiac Intelligence Corporation | Automated system and method for establishing a patient status reference baseline |
| 7257438, | Jul 23 2002 | Datascope Investment Corp. | Patient-worn medical monitoring device |
| 7277752, | Jun 11 2003 | System for cardiac resuscitation | |
| 7294108, | Jan 27 2005 | Pacesetter, Inc. | Cardiac event microrecorder and method for implanting same |
| 7328061, | Dec 29 2000 | GE Medical Systems Information Technologies | Method and apparatus for perioperative assessment of cardiovascular risk |
| 7412395, | Dec 29 2000 | GE MEDICAL SYSTEMS INFORMATION TECHNOLOGIES, INC | Automated scheduling of emergency procedure based on identification of high-risk patient |
| 7429938, | Mar 26 2007 | Medtronic, Inc. | Method of compressing waveform data with differential entropy based compression |
| 7552031, | Dec 04 2002 | Apple Inc | Personal items network, and associated methods |
| 7706870, | Jul 10 2007 | Yuan Ze University | Method for analyzing irreversible apneic coma (IAC) |
| 7756721, | Mar 14 1997 | BEST DOCTORS, INC ; TELADOC HEALTH, INC [F K A TELADOC, INC ] | Health care management system |
| 7787943, | Jul 25 2007 | Heart rate monitor for swimmers | |
| 7874993, | Nov 16 1999 | Cardiac Pacemakers, Inc. | System and method for diagnosing and monitoring congestive heart failure |
| 7881785, | Mar 26 2008 | ZOLL Medical Corporation | Method and apparatus for defrosting a defibrillation electrode |
| 7959574, | Nov 16 1999 | Cardiac Pacemakers, Inc. | System and method for analyzing a patient status for respiratory insufficiency for use in automated patient care |
| 8108035, | Oct 18 2006 | Pacesetter, Inc. | Systems and methods for detecting and compensating for changes in posture during ischemia detection a using an implantable medical device |
| 8116841, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device with multiple physiological sensors |
| 8135459, | Sep 18 2000 | Cameron Health, Inc. | Unitary subcutaneous only implantable cardioverter-defibrillator and optional pacer |
| 8172761, | Sep 28 2004 | YUKKA MAGIC LLC | Monitoring device with an accelerometer, method and system |
| 8180425, | Dec 05 2006 | KPR U S , LLC | ECG lead wire organizer and dispenser |
| 8200320, | Mar 03 2006 | Physiowave, Inc.; PHYSIOWAVE INC | Integrated physiologic monitoring systems and methods |
| 8231539, | Nov 16 1999 | Cardiac Pacemakers, Inc. | Automated patient care system and method for diagnosing and monitoring respiratory insufficiency |
| 8231540, | Nov 16 1999 | Cardiac Pacemakers, Inc. | System and method for managing respiratory insufficiency in conjunction with heart failure assessment |
| 8239012, | Oct 08 2010 | ZOLL Medical Corporation | Microcontrolled electrocardiographic monitoring circuit with differential voltage encoding |
| 8249686, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device for sleep disordered breathing |
| 8260414, | Mar 26 2008 | ZOLL Medical Corporation | Method and apparatus for defrosting a defibrillation electrode |
| 8266008, | Aug 15 2006 | Amazon Technologies, Inc. | Facilitating a supply of used items |
| 8277378, | Jun 03 1999 | Cardiac Pacemakers, Inc | System and method for collection and analysis of patient information for automated remote patient care |
| 8285356, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device with multiple physiological sensors |
| 8285370, | Oct 08 2010 | ZOLL Medical Corporation | Microcontrolled electrocardiographic monitoring circuit with feedback control |
| 8308650, | Jul 26 1999 | Cardiac Pacemakers, Inc. | System and method for generating baseline data for automated management of cardiovascular pressure |
| 8366629, | Nov 16 1999 | Cardiac Pacemakers, Inc. | System and method for diagnosing and monitoring congestive heart failure |
| 8374688, | Sep 14 2007 | MEDTRONIC MONITORING, INC | System and methods for wireless body fluid monitoring |
| 8412317, | Apr 18 2008 | MEDTRONIC MONITORING, INC | Method and apparatus to measure bioelectric impedance of patient tissue |
| 8460189, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent cardiac monitor with advanced sensing capabilities |
| 8473047, | Aug 03 2010 | MEDTRONIC MONITORING, INC | Multifrequency bioimpedence device and related methods |
| 8478418, | Apr 15 2011 | INFOBIONIC, INC | Remote health monitoring system |
| 8538503, | May 12 2010 | iRhythm Technologies, Inc. | Device features and design elements for long-term adhesion |
| 8554311, | Jun 17 2011 | General Electric Company | System and method of noise reduction in an electrocardiology study |
| 8560046, | May 12 2010 | iRhythm Technologies, Inc | Device features and design elements for long-term adhesion |
| 8591430, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device for respiratory monitoring |
| 8594763, | May 25 2010 | NeuroWave Systems Inc.; NeuroWave Systems Inc | Physiological electrode assembly for fast application |
| 8600486, | Mar 25 2011 | ZOLL Medical Corporation | Method of detecting signal clipping in a wearable ambulatory medical device |
| 8613708, | Oct 08 2010 | ZOLL Medical Corporation | Ambulatory electrocardiographic monitor with jumpered sensing electrode |
| 8613709, | Oct 08 2010 | ZOLL Medical Corporation | Ambulatory electrocardiographic monitor for providing ease of use in women |
| 8620418, | Jan 04 2013 | InfoBionic, Inc. | Systems and methods for processing and displaying patient electrocardiograph data |
| 8626277, | Oct 08 2010 | ZOLL Medical Corporation | Computer-implemented electrocardiographic data processor with time stamp correlation |
| 8628020, | Oct 24 2007 | HMICRO | Flexible wireless patch for physiological monitoring and methods of manufacturing the same |
| 8668653, | Mar 24 2004 | NIHON KOHDEN CORPORATION | Biological information measuring garment having sensor, biological information measuring system and equipment, and control method of equipment |
| 8684925, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Injectable device for physiological monitoring |
| 8688190, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device for sleep disordered breathing |
| 8718752, | Mar 12 2008 | MEDTRONIC MONITORING, INC | Heart failure decompensation prediction based on cardiac rhythm |
| 8744561, | Apr 15 2011 | InfoBionic, Inc. | Remote health monitoring system |
| 8774932, | Apr 15 2011 | InfoBionic, Inc. | Remote health monitoring system |
| 8790257, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Multi-sensor patient monitor to detect impending cardiac decompensation |
| 8790259, | Oct 22 2009 | MEDTRONIC MONITORING, INC | Method and apparatus for remote detection and monitoring of functional chronotropic incompetence |
| 8795174, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device with multiple physiological sensors |
| 8798729, | Mar 25 2011 | ZOLL Medical Corporation | Method of detecting signal clipping in a wearable ambulatory medical device |
| 8798734, | Jan 04 2013 | Infobionic Inc. | Systems and methods for processing and displaying patient electrocardiograph data |
| 8818478, | Mar 31 2011 | adidas AG | Sensor garment |
| 8818481, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device with multiple physiological sensors |
| 8823490, | Dec 15 2008 | MEDTRONIC MONITORING, INC | Patient monitoring systems and methods |
| 8938287, | Oct 08 2010 | ZOLL Medical Corporation | Computer-implemented electrocardiograhic data processor with time stamp correlation |
| 8965492, | Nov 01 2006 | FITLINXX, INC | Body worn physiological sensor device having a disposable electrode module |
| 9066664, | Mar 13 2009 | Polar Electro Oy | Data transfer |
| 9155484, | Nov 01 2006 | FITLINXX, INC | Body worn physiological sensor device having a disposable electrode module |
| 9204813, | Mar 25 2011 | ZOLL Medical Corporation | Method of detecting signal clipping in a wearable ambulatory medical device |
| 9241649, | May 12 2010 | iRhythm Technologies, Inc. | Device features and design elements for long-term adhesion |
| 9259154, | Sep 27 2005 | SENSIUM HEALTHCARE LIMITED | Monitoring method and apparatus |
| 9277864, | May 24 2012 | VITAL CONNECT, INC. | Modular wearable sensor device |
| 9339202, | Nov 03 2009 | VIVAQUANT, INC | System for processing physiological data |
| 9375179, | Dec 23 2010 | Biosense Webster, Inc | Single radio-transparent connector for multi-functional reference patch |
| 9414786, | Nov 03 2009 | VIVAQUANT, INC | ECG sensing with noise filtering |
| 9439566, | Mar 15 2013 | OTSUKA PHARMACEUTICAL CO , LTD | Re-wearable wireless device |
| 9597004, | Oct 31 2014 | iRhythm Technologies, Inc | Wearable monitor |
| 9603542, | Jul 13 2009 | Koninklijke Philips Electronics N V | Electro-physiological measurement with reduced motion artifacts |
| 9700222, | Dec 02 2011 | LumiraDx UK Ltd | Health-monitor patch |
| 9770182, | Sep 14 2007 | MEDTRONIC MONITORING, INC | Adherent device with multiple physiological sensors |
| 20020013538, | |||
| 20020013717, | |||
| 20020016798, | |||
| 20020103422, | |||
| 20020109621, | |||
| 20020120310, | |||
| 20020128686, | |||
| 20020184055, | |||
| 20020193668, | |||
| 20030004547, | |||
| 20030028811, | |||
| 20030073916, | |||
| 20030083559, | |||
| 20030097078, | |||
| 20030139785, | |||
| 20030176802, | |||
| 20030211797, | |||
| 20040008123, | |||
| 20040019288, | |||
| 20040034284, | |||
| 20040049120, | |||
| 20040049132, | |||
| 20040073127, | |||
| 20040087836, | |||
| 20040088019, | |||
| 20040093192, | |||
| 20040116784, | |||
| 20040148194, | |||
| 20040163034, | |||
| 20040167416, | |||
| 20040207530, | |||
| 20040210165, | |||
| 20040236202, | |||
| 20040243435, | |||
| 20040256453, | |||
| 20040260188, | |||
| 20040260192, | |||
| 20050010139, | |||
| 20050096717, | |||
| 20050108055, | |||
| 20050151640, | |||
| 20050154267, | |||
| 20050182308, | |||
| 20050182309, | |||
| 20050215918, | |||
| 20050222513, | |||
| 20050228243, | |||
| 20050245839, | |||
| 20050261564, | |||
| 20050275416, | |||
| 20060025696, | |||
| 20060025824, | |||
| 20060030767, | |||
| 20060030904, | |||
| 20060041201, | |||
| 20060054737, | |||
| 20060084883, | |||
| 20060100530, | |||
| 20060111642, | |||
| 20060122469, | |||
| 20060124193, | |||
| 20060224072, | |||
| 20060229522, | |||
| 20060235320, | |||
| 20060253006, | |||
| 20060264730, | |||
| 20060264767, | |||
| 20070003115, | |||
| 20070038057, | |||
| 20070050209, | |||
| 20070078324, | |||
| 20070078354, | |||
| 20070088406, | |||
| 20070089800, | |||
| 20070093719, | |||
| 20070100248, | |||
| 20070100667, | |||
| 20070123801, | |||
| 20070131595, | |||
| 20070136091, | |||
| 20070179357, | |||
| 20070185390, | |||
| 20070203415, | |||
| 20070203423, | |||
| 20070208232, | |||
| 20070208233, | |||
| 20070208266, | |||
| 20070225611, | |||
| 20070244405, | |||
| 20070249946, | |||
| 20070255153, | |||
| 20070265510, | |||
| 20070276270, | |||
| 20070276275, | |||
| 20070293738, | |||
| 20070293739, | |||
| 20070293740, | |||
| 20070293741, | |||
| 20070293772, | |||
| 20070299325, | |||
| 20070299617, | |||
| 20080027339, | |||
| 20080051668, | |||
| 20080058661, | |||
| 20080088467, | |||
| 20080091089, | |||
| 20080091097, | |||
| 20080108890, | |||
| 20080114232, | |||
| 20080139953, | |||
| 20080143080, | |||
| 20080177168, | |||
| 20080194927, | |||
| 20080208009, | |||
| 20080208014, | |||
| 20080284599, | |||
| 20080288026, | |||
| 20080294024, | |||
| 20080306359, | |||
| 20080312522, | |||
| 20090012412, | |||
| 20090012979, | |||
| 20090054952, | |||
| 20090062897, | |||
| 20090069867, | |||
| 20090073991, | |||
| 20090076336, | |||
| 20090076341, | |||
| 20090076342, | |||
| 20090076343, | |||
| 20090076346, | |||
| 20090076349, | |||
| 20090076397, | |||
| 20090076401, | |||
| 20090076559, | |||
| 20090088652, | |||
| 20090112116, | |||
| 20090131759, | |||
| 20090156908, | |||
| 20090216132, | |||
| 20090270708, | |||
| 20090270747, | |||
| 20090292194, | |||
| 20100007413, | |||
| 20100022897, | |||
| 20100056881, | |||
| 20100081913, | |||
| 20100174229, | |||
| 20100177100, | |||
| 20100185063, | |||
| 20100185076, | |||
| 20100191154, | |||
| 20100191310, | |||
| 20100223020, | |||
| 20100234715, | |||
| 20100234716, | |||
| 20100280366, | |||
| 20100312188, | |||
| 20100324384, | |||
| 20110021937, | |||
| 20110054286, | |||
| 20110060215, | |||
| 20110066041, | |||
| 20110077497, | |||
| 20110105861, | |||
| 20110144470, | |||
| 20110160548, | |||
| 20110224564, | |||
| 20110237922, | |||
| 20110237924, | |||
| 20110245699, | |||
| 20110245711, | |||
| 20110288605, | |||
| 20120003933, | |||
| 20120029306, | |||
| 20120029315, | |||
| 20120029316, | |||
| 20120035432, | |||
| 20120059668, | |||
| 20120078127, | |||
| 20120088998, | |||
| 20120088999, | |||
| 20120089000, | |||
| 20120089001, | |||
| 20120089037, | |||
| 20120089412, | |||
| 20120089417, | |||
| 20120095352, | |||
| 20120101358, | |||
| 20120101396, | |||
| 20120165645, | |||
| 20120172695, | |||
| 20120220835, | |||
| 20120232929, | |||
| 20120238910, | |||
| 20120253847, | |||
| 20120302906, | |||
| 20120306662, | |||
| 20120330126, | |||
| 20130041272, | |||
| 20130077263, | |||
| 20130079611, | |||
| 20130085347, | |||
| 20130085403, | |||
| 20130087609, | |||
| 20130096395, | |||
| 20130116533, | |||
| 20130123651, | |||
| 20130158361, | |||
| 20130197380, | |||
| 20130225963, | |||
| 20130225966, | |||
| 20130231947, | |||
| 20130243105, | |||
| 20130274584, | |||
| 20130275158, | |||
| 20130324809, | |||
| 20130324855, | |||
| 20130324856, | |||
| 20130325081, | |||
| 20130325359, | |||
| 20130331665, | |||
| 20130338448, | |||
| 20130338472, | |||
| 20140012154, | |||
| 20140056452, | |||
| 20140088399, | |||
| 20140107509, | |||
| 20140140359, | |||
| 20140180027, | |||
| 20140189928, | |||
| 20140206977, | |||
| 20140214134, | |||
| 20140215246, | |||
| 20140249852, | |||
| 20140296651, | |||
| 20140343390, | |||
| 20140358193, | |||
| 20140364756, | |||
| 20150048836, | |||
| 20150065842, | |||
| 20150164349, | |||
| 20150165211, | |||
| 20150177175, | |||
| 20150250422, | |||
| 20150257670, | |||
| 20150305676, | |||
| 20150359489, | |||
| 20160135746, | |||
| 20160144190, | |||
| 20160144192, | |||
| 20160217691, | |||
| 20160235318, | |||
| 20170112399, | |||
| 20170156592, | |||
| 20170281032, | |||
| 20170366921, | |||
| 20180078771, | |||
| 20190021671, | |||
| 20190059763, | |||
| D341423, | Feb 14 1991 | Caliber Medical Corporation | Ambulatory cardiac monitor |
| D357069, | Aug 04 1993 | Cardiac Science Corporation | Medical electrode |
| D377983, | Sep 13 1995 | CARD GUARD TECHNOLOGIES, INC | Cardiac monitor |
| D407159, | Apr 30 1998 | Pre-natal heartbeat monitor | |
| D443063, | Aug 25 2000 | CLEARPATH PARTNERS, LLC | Chest multisensor array |
| D445507, | Aug 25 2000 | CLEARPATH PARTNERS, LLC | Electronics unit for chest multisensor array |
| D558882, | Mar 14 2006 | AMBU A S | Biomedical electrode for attachment to skin |
| D606656, | Apr 04 2008 | Seiko Epson Corporation | Wrist watch type purse sensor |
| D639437, | Oct 08 2010 | ZOLL Medical Corporation | Wearable ambulatory electrocardiographic monitor |
| DE19955211, | |||
| EP1859833, | |||
| EP2438851, | |||
| EP2438852, | |||
| EP2465415, | |||
| EP2589333, | |||
| JP11188015, | |||
| JP2004129788, | |||
| JP2007082938, | |||
| JP2009219554, | |||
| JP6319711, | |||
| WO78213, | |||
| WO199852463, | |||
| WO2003032192, | |||
| WO2006009767, | |||
| WO2006014806, | |||
| WO2007066270, | |||
| WO2007092543, | |||
| WO2008010216, | |||
| WO2008057884, | |||
| WO2008092098, | |||
| WO2009036306, | |||
| WO2009036313, | |||
| WO2009036327, | |||
| WO2009112976, | |||
| WO2009112978, | |||
| WO2009112979, | |||
| WO2009142975, | |||
| WO2010066507, | |||
| WO2010105045, | |||
| WO2011047207, | |||
| WO2012140559, | |||
| WO2012146957, | |||
| WO2017072250, |
| Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
| Nov 20 2018 | BOLEYN, RODNEY | BARDY DIAGNOSTICS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 047584 | /0024 | |
| Nov 23 2018 | DREISBACH, EZRA M | BARDY DIAGNOSTICS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 047584 | /0024 | |
| Nov 23 2018 | DULKEN, CHUCK | BARDY DIAGNOSTICS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 047584 | /0024 | |
| Nov 26 2018 | BARDY, GUST H | BARDY DIAGNOSTICS, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 047584 | /0024 | |
| Nov 26 2018 | BARDY DIAGNOSTICS, INC. | (assignment on the face of the patent) | / | |||
| Oct 22 2021 | BARDY DIAGNOSTICS, INC | JPMORGAN CHASE BANK, N A | SECURITY AGREEMENT SUPPLEMENT | 058567 | /0677 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | Hill-Rom, Inc | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | VOALTE, INC | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | Welch Allyn, Inc | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | HILL-ROM HOLDINGS, INC | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | ALLEN MEDICAL SYSTEMS, INC | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | Hill-Rom Services, Inc | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | BREATHE TECHNOLOGIES, INC | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 | |
| Dec 13 2021 | JPMORGAN CHASE BANK, N A | BARDY DIAGNOSTICS, INC | RELEASE OF SECURITY INTEREST SENT FOR RECORDAL OCTOBER 25, 2021 | 058516 | /0312 |
| Date | Maintenance Fee Events |
| Nov 26 2018 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
| Dec 14 2018 | SMAL: Entity status set to Small. |
| Oct 27 2021 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
| Apr 24 2023 | M1551: Payment of Maintenance Fee, 4th Year, Large Entity. |
| Date | Maintenance Schedule |
| Nov 05 2022 | 4 years fee payment window open |
| May 05 2023 | 6 months grace period start (w surcharge) |
| Nov 05 2023 | patent expiry (for year 4) |
| Nov 05 2025 | 2 years to revive unintentionally abandoned end. (for year 4) |
| Nov 05 2026 | 8 years fee payment window open |
| May 05 2027 | 6 months grace period start (w surcharge) |
| Nov 05 2027 | patent expiry (for year 8) |
| Nov 05 2029 | 2 years to revive unintentionally abandoned end. (for year 8) |
| Nov 05 2030 | 12 years fee payment window open |
| May 05 2031 | 6 months grace period start (w surcharge) |
| Nov 05 2031 | patent expiry (for year 12) |
| Nov 05 2033 | 2 years to revive unintentionally abandoned end. (for year 12) |